The goal of this project is to define the molecular mechanisms involved in the replication of HIV and related retroviruses and to develop new strategies for AIDS therapy. Our research is currently focused on several broad areas of interest: reverse transcription; inhibition of virus replication by the host protein APOBEC3G; and virus assembly. During reverse transcription, there are two strand transfer events that are required for synthesis of full-length plus- and minus-strand DNA copies of the viral RNA genome. This process is dependent on the viral nucleocapsid protein (NC), a nucleic acid chaperone with the ability to catalyze conformational rearrangements that lead to the most thermodynamically stable nucleic acid structures. HIV-1 NC has two zinc fingers, each containing the invariant CCHC zinc-coordinating motifs. (A) Using a mutational approach, we have provided the first direct evidence that zinc coordination by the CCHC motifs is required for efficient strand transfer. The zinc fingers are also required for NC-mediated inhibition of dead-end self-priming reactions, which compete with minus-strand transfer and are induced by the complementary TAR secondary structure at the 3' end of (-) strong-stop DNA. In recent work on NC, we have focused on two issues that affect minus-strand transfer. (i) To investigate the effect of changes in nucleic acid structure and thermostability on NC-mediated minus-strand transfer, we designed a series of (-) strong-stop DNA and acceptor RNA constructs. Biochemical data were correlated with enzymatic mapping and mFold analysis. We found that eliminating self-priming by disrupting the TAR DNA structure does not necessarily result in increased strand transfer efficiency: acceptor RNA structure is also critical. Thus, strand transfer is efficient only when (-) strong-stop DNA and acceptor RNA are moderately structured and a delicate thermodynamic balance between these two reactants is maintained. More recently, we have obtained evidence demonstrating that NC nucleic acid chaperone activity is ultimately dependent on the stability of local structure at the nucleation site for annealing, rather than on the overall stability of the structure. (ii) Short 5' terminal RNA fragments generated during RNase H degradation of genomic RNA are initially annealed to the 3' end of (-) SSDNA. These fragments must be removed so that strand transfer can occur. By modeling this reaction in the context of minus-strand transfer, we have shown that fragment removal can be catalyzed by NC alone without a requirement for secondary RNase H activity (as previously thought). Coordination of zinc by the CCHC motifs is required. (B) Other studies are being performed on the initiation step in HIV-1 reverse transcription. This event is primed by a host lysyl-tRNA, which is annealed to the 18-nt primer-binding site near the 5' terminus of the viral RNA genome; extension of the primer leads to synthesis of (-) strong-stop DNA. Mutational analysis and band-shift experiments with wild-type and mutant template RNAs support the possibility that NC promotes an interaction between the 3' arm of the anticodon stem and part of the variable loop of the tRNA primer and nt 143-149 in viral RNA. Zinc coordination is not required to facilitate this interaction. (C) We have recently initiated a new project on APOBEC3G (APO3G), a cellular cytidine deaminase with two zinc finger domains, which blocks HIV-1 reverse transcription and replication in the absence of the viral protein known as Vif. We have succeeded in purifying catalytically active APO3G, allowing us for the first time, to provide a comprehensive molecular analysis of its deaminase and nucleic acid binding activities. We have demonstrated that APO3G deaminates cytosines in single-stranded (ss)DNA only, whereas it binds efficiently to ssDNA and ssRNA. Moreover, when bound to an 18-nt ssDNA, APO3G forms a dimer. Although both zinc finger domains have the ability to bind nucleic acids, the first zinc finger contributes more to binding and APO3G encapsidation into virions than finger two. In contrast, deamination is associated exclusively with the second zinc finger. Moreover, zinc finger two is more important than finger one for the antiviral effect, demonstrating a correlation between deaminase and antiviral activities. Experiments to elucidate the mechanism(s) by which APO3G inhibits HIV-1 reverse transcription are the focus of current efforts. (D) Our laboratory has also been investigating the role of the HIV-1 capsid protein (CA) in early postentry events, a stage in the infectious process that is still not completely understood. (i) We have previously described the unusual phenotype associated with single alanine substitution mutations in conserved N-terminal hydrophobic residues. Mutant virions are not infectious and lack a cone-shaped core. Moreover, despite having a functional RT protein, the mutants are blocked in viral DNA synthesis in cells, indicating a defect in an early postentry event preceding reverse transcription. Additionally, mutant cores have a marked deficiency in RT protein and enzymatic activity and a substantial increase in the retention of CA. The high level of core-associated CA would interfere with proper disassembly and taken together with the RT defect, would also account for the failure of the mutants to synthesize viral DNA postentry. Our results demonstrate the intimate connection between infectivity, proper core morphology, and the ability to undergo reverse transcription. (ii) We have also performed studies to determine whether substitutions other than alanine result in a different phenotype. A total of 13 substitutions were made at position 23 (Trp), but only one mutant, W23F, was infectious, albeit to a very small extent. W23F, unlike the original mutant W23A, was able to replicate during long-term culture in MT-4 cells, but with delayed replication kinetics. With continued passage, we eventually isolated two second-site suppressor mutants, which replicate like wild-type virus in MT-4 cells. We have also found that the original W23A mutant has trans-dominant inhibitory activity i.e., W23A can reduce the infectivity of wild-type virus produced by cells cotransfected with a mixture of wild-type and mutant particles. Interestingly, the trans-dominant inhibitory activity of W23F is approximately five-fold less than the corresponding activity of W23A. These observations indicate that the virions resulting from the cotransfection contain a co-assembled population of wild-type and mutant CA molecules. Moreover, the results indicate that collaboration of many N-terminal CA domains is required for assembly of a virus particle, in accord with models for CA multimerization. (iii) In current work, we are investigating the effect of point mutations in the linker region that connects the N- and C-terminal domains of CA.